Implement int8 symmetric per-channel weight quantization and dequantization for a linear l

Medium

Prefill, Decode, and KV Cache Memory2 / 5

Chapter 022 / 5

Prefill, Decode, and KV Cache Memory

Implement int8 symmetric per-channel weight quantization and dequantization for a linear l

Medium

Implement int8 symmetric per-channel weight quantization and dequantization for a linear layer.

Target interface

quantize_per_channel(W)

Complete the starter code in the editor. Any correct approach is accepted.

Hints0 / 2

Reference solution

Your own code stays in the editor.

solution.pyPython

Local draft

import numpy as np

def quantize_per_channel(W):
    raise NotImplementedError

def dequantize_per_channel(q, scale):
    raise NotImplementedError

def quantized_linear(x, q, scale, bias=None):
    raise NotImplementedError

Local draft

⌘/Ctrl↵to submit

AI review

Ready when you are

Submit your solution for a score and feedback on correctness, edge cases, and clarity. Any correct approach is accepted.

Chapter 022 / 5

Prefill, Decode, and KV Cache Memory

Implement int8 symmetric per-channel weight quantization and dequantization for a linear l

Medium

Implement int8 symmetric per-channel weight quantization and dequantization for a linear layer.

Target interface

quantize_per_channel(W)

Complete the starter code in the editor. Any correct approach is accepted.

Hints0 / 2

Reference solution

Your own code stays in the editor.